b10331 b10330
ggml-org/llama.cppAug 8, 2026 → Aug 8, 20260 days between releases
Generating AI comparison
Analyzing what changed between b10331 and b10330. This page updates automatically — usually within a minute.
The Two Releases
b10331
b10331
Aug 8, 2026· 2 months ago
Fixes the `get_info` API response to correctly report the working directory when tools are running in an isolated environment.
Full release notesb10330
b10330
Aug 8, 2026· 2 months ago
Optimizes CUDA performance by fusing RMS normalization, multiplication, and RoPE operations into a single kernel launch.
Full release notes